Mechanisms linking metabolism of Helicobacter pylori to 18O and 13C-isotopes of human breath CO2

The gastric pathogen Helicobacter pylori utilize glucose during metabolism, but the underlying mechanisms linking to oxygen-18 (18O) and carbon-13 (13C)-isotopic fractionations of breath CO2 during glucose metabolism are poorly understood. Using the excretion dynamics of 18O/16O and 13C/12C-isotope ratios of breath CO2, we found that individuals with Helicobacter pylori infections exhibited significantly higher isotopic enrichments of 18O in breath CO2 during the 2h-glucose metabolism regardless of the isotopic nature of the substrate, while no significant enrichments of 18O in breath CO2 were manifested in individuals without the infections. In contrast, the 13C-isotopic enrichments of breath CO2 were significantly higher in individuals with Helicobacter pylori compared to individuals without infections in response to 13C-enriched glucose uptake, whereas a distinguishable change of breath 13C/12C-isotope ratios was also evident when Helicobacter pylori utilize natural glucose. Moreover, monitoring the 18O and 13C-isotopic exchange in breath CO2 successfully diagnosed the eradications of Helicobacter pylori infections following a standard therapy. Our findings suggest that breath 12C18O16O and 13C16O16O can be used as potential molecular biomarkers to distinctively track the pathogenesis of Helicobacter pylori and also for eradication purposes and thus may open new perspectives into the pathogen’s physiology along with isotope-specific non-invasive diagnosis of the infection.

complete understanding of glucose metabolism during the H. pylori infection could be of significance in the development of novel therapies for the micro-organism alongside new and better approaches for treating the most common human bacterial infection.
Furthermore, an earlier study revealed that the oxygen-16 ( 16 O) and the oxygen-18 ( 18 O) isotopes in 12 C 16 O 2 and water (H 2 18 O), respectively, are rapidly interchanged during the human respiration process mediated by the metalloenzyme carbonic anhydrase (CA) 10,11 . It is also known that H. pylori encodes two different forms of the metalloenzyme carbonic anhydrase (α -CA and β -CA) 12 and this gastric pathogen plays a vital role in inter-conversion of carbon dioxide and bicarbonate (CO 2 + H 2 O  H + + HCO 3 − ), catalyzed by the CA activity [12][13][14] . This efficient activity suggests that investigations of breath 18 O/ 16 O isotopic fractionations of CO 2 may specifically track the gastric pathogen H. pylori and hence may introduce a novel non-invasive strategy in the diagnosis of H. pylori infections living in human stomach. As a consequence, we hypothesized that simultaneous monitoring the 18 O/ 16 O and 13 C/ 12 C stable isotope ratios of exhaled breath CO 2 associated with glucose metabolism in H. pylori may act as potential markers for the early detection of H. pylori infections or during the preclinical phase of the infections. In view of the fact that H. pylori is able to uptake and metabolize glucose confirmed as experimentally 15 and also by analysing the genome sequence 5 therefore, there is a pressing need to assess the clinical efficacy of the glucose utilization by H. pylori for large-scale screening individuals harboring the micro-organism. In addition, unravelling the precise metabolic pathways involved in causing the isotopic fractionations of 12 16 O in human breath during the glucose uptake by H. pylori remains a major challenge, whenever an individual is at-risk of developing the disease.
In this study, we first report, the potential links of both 18 O and 13 C-stable isotopes of breath CO 2 with the gastric pathogen H. pylori in response to glucose ingestion. We investigated simultaneously the time-dependent excretion dynamics of the 12 16 O isotope ratios of breath CO 2 from individuals with H. pylori positive and negative by employing a laser-based integrated cavity output spectroscopy (ICOS) method. We further explored the potential metabolic pathways underlying the glucose utilization in the pathogenesis of H. pylori infection and the mechanisms linking breath oxygen-18 and carbon-13 isotopic fractionations of CO 2 to the gastric pathogen H. pylori. Finally, we determined various diagnostic parameters such as optimal diagnostic cut-off values, diagnostic sensitivity and specificity of oxygen-18 and carbon-13 stable isotopes in breath CO 2 to gain a better insight into the diagnostic efficiency for the non-invasive detection of H. pylori infection in real-time.

Results and Discussion
To investigate the 18 O and 13 C isotopic fractionations of breath CO 2 , we first studied the time-dependent excretion dynamics of both isotopes in exhaled breath after ingestion of an oral dose of 13 C-enriched glucose for H. pylori positive (n = 72) and negative (n = 55) individuals, using a laser-based high-precision cavity-enhanced integrated cavity output spectrometer (ICOS). We explored the isotopic fractionation of CO 2 by simultaneous monitoring the 18 O/ 16 O and 13 C/ 12 C stable isotope ratios in breath, expressed as delta-over-baseline (DOB) relative to the Vienna Pee Dee Belemnite standard, i.e., δ DOB 18 O‰ = [(δ 18 O‰ ) t=t -(δ 18 O‰) t=basal ] and δ DOB 13 C‰ = [(δ 13 C‰) t=t -(δ 13 C‰) t=basal ], respectively, associated with glucose metabolism. In this investigation (Fig. 1a), individuals with H. pylori positive exhibited significantly higher isotopic enrichments of 18 O in CO 2 compared with H. pylori negative during the 2h-glucose metabolism, while no significant enrichments of 18 O in CO 2 were manifested in individuals without H. pylori infections. These findings suggest a potential link between H. pylori infections in human stomach and the 18 O-isotopic fractionations in exhaled breath and hence may open a new route for the non-invasive assessment of H. pylori infections. Carbonic anhydrase (CA) activity of H. pylori has previously been proposed to interchange the oxygen isotopes of CO 2 ( 16 O) and H 2 O ( 18 O) efficiently 10,11 , suggesting that in our observations CA activity may play an important role in oxygen-isotope fractionations of breath CO 2 in the glucose-mediated bacterial environment. Hence a statistically significant difference in the δ DOB 18 O‰ values in excretion dynamics established a marked distinction (Fig. 1b) between H. pylori infected and non-infected individuals. Taken together, these findings indicate that the monitoring of 18 O-exchange between C 16 O 2 and H 2 18 O in response to CA activity may distinctively track the development of the gastric pathogen in human stomach and might be considered as a potential biomarker for the non-invasive detection of H. pylori infection.
We then critically assessed the excretion dynamics of δ DOB 13 C (‰) (Fig. 1a) values in exhaled breath samples in response to 13 C-enriched glucose ingestion. The isotopic enrichments of 13 C in breath CO 2 were significantly higher (Fig. 1c) in individuals with H. pylori positive compared with individuals with H. pylori negative. It was noticed that for H. pylori positive patients the δ DOB 13 C (‰) values increased gradually with time at a faster rate in comparison with individuals without H. pylori infections. Several lines of evidence suggest that H. pylori can metabolize glucose in both oxidative and fermentative pathways 8,9 and as a consequence the catabolism of glucose resulted in higher isotopic enrichments of 13 C in exhaled breath CO 2 . Moreover, in some early evidences 9,16 , it was demonstrated the biphasic characteristics of glucose utilization by H. pylori with a slower initial period, followed by a second phase with a higher rate of glucose uptake. The transport and utilization of glucose was previously investigated into the intact micro-organism employing the radioactive tracer analysis. Therefore, the gradual increase of the δ DOB  16 , where the uptake of glucose into H. pylori cells exhibited the biphasic patterns. Our observations therefore, point to new perspectives into the physiology of H. pylori underlying the isotopic fractionations of 13 C in breath CO 2 associated with glucose metabolism.
We next explored how the time-dependent excretion dynamics of isotopic breath CO 2 changes after administration of unlabelled glucose (i.e. with no 13 C-enriched glucose), as the potential role of glucose metabolism in response to unlabelled glucose ingestion for individuals with H. pylori infection and the possible links underlying the 18  O‰ values in breath samples over time was evident for H. pylori negative individuals (Fig. 2a). These findings suggest that the mechanisms i.e. oxidation of glucose in the bacterial environment to form bicarbonate (HCO 3 − ) and subsequently the enzyme carbonic anhydrase-mediated rapid inter-conversion of HCO 3 − and CO 2 , leading to the generation of 12 C 16 O 18 O, exclusively depends on the substrate (glucose) regardless of its isotopic nature. Interestingly, although the isotopic nature of the substrate is vital to observe effectively the 13 C-isotopic enrichments of breath CO 2 (i.e. enhancement of δ DOB 13 C‰ values), yet the enrichments of δ DOB 13 C‰ values, due to natural abundances of 13 C, during glucose metabolism of H. pylori are significantly distinguishable for H. pylori positive patients (Fig. 2c,d), suggesting a new step towards characterizing the transport and utilization of unlabelled glucose into the human pathogen for better understanding of its physiology in the gastric niche. In view of the breath results, we have also established the previous hypothesis 12 that the bacterium requires high CO 2 for growth and the interconversion of 18 Table 1.
To distinctively track the H. pylori infection as well as for early detection prior to the onset of different gastric diseases, we subsequently determined numerous statistically sound optimal diagnostic cut-off points of δ DOB 18 O‰ and δ DOB 13 C‰ values in exhaled breath associated with 13 C-labelled and unlabelled glucose metabolism, using receiver operating characteristics curve (ROC) analysis (Fig. 3). O‰ ≥ 1.1‰ were considered to be H. pylori positive with and without 13 C-enriched glucose metabolism respectively, and these corresponded to the diagnostic sensitivity and specificity of 100% and ~98%, respectively. On the contrary, a different optimal diagnostic cut-off point of δ DOB 13 C‰ ≥ 33.32‰ between individuals with H. pylori positive and negative,  demonstrated the sensitivity and specificity of 100% and 100%, respectively, when 13 C-labelled glucose is ingested, whereas without 13 C-labelled glucose, δ DOB 13 C‰ ≥ 1.51‰ precisely diagnosed the infected and non-infected persons corresponding to the similar levels of diagnostic sensitivity (100%) and specificity (98%). It is noteworthy to mention that the uncertainty of these cut-off values is associated with the less-sensitive techniques for isotope measurements and the variation of isotopic fractionations in the test meal. However, these findings suggest that the oxygen-18 and carbon-13 isotopic fractionations of the major metabolite CO 2 in human breath linked to glucose metabolism of H. pylori provide a new non-invasive approach to treat the world's most common chronic bacterial infection of humans and hence may have a broad clinical efficacy for precise assessment of the gastric pathogen H. pylori.
We next explored the efficacy of the glucose breath test in response to the standard eradication therapies of the infection. A marked depletions of both δ DOB 18 O‰ and δ DOB 13 C‰ values for H. pylori infected patients (n = 37 for 13 C-glucose and n = 28 for 12 C-glucose) (Fig. 4) after complete eradication of the infection were manifested, suggesting the widespread clinical significance of the glucose breath test. Our findings associated with the glucose metabolism by H. pylori infections thus point towards a considerable clinical advancement in the non-invasive diagnosis of H. pylori infection by contrast with the currently available 13 C-urea breath test ( 13 C-UBT), where 13 C-enriched substrate (urea) is usually used. In view of this result, we therefore posit that the glucose breath test by ingestion of a natural substrate (unlabelled glucose) is a valid and potentially robust new-generation diagnostic tool and thus indicate great promise for comparatively less-expensive and non-toxic global technique, in comparison with the 13 C-UBT, for the non-invasive assessment i.e. early detection and follow-up of patients after eradication of H. pylori infection.
Finally, we elucidated the potential metabolic pathways (Fig. 5) underlying the mechanisms linking isotopic fractionations of breath CO 2 and glucose utilization by H. pylori infection. When a dose of glucose is orally administered to the patients, the ingested glucose disposal takes place in the cytoplasm of H. pylori through the HP1174 transporter (protein) 7 . After glucose enters into the cytoplasm, it is phosphorylated to produce glucose-6-phosphate which subsequently incorporated with three potential metabolic routes: glycolysis, pentose phosphate and the Entner-Doudoroff pathway 17,18 . A part of the total glucose-6-phosphate, which goes into the pentose phosphate pathway, is predominantly oxidised into CO 2 . The remaining part of glucose-6-phosphate enters into the other two metabolic pathways and may lead to the generation of pyruvate 17 and eventually gives rise to CO 2 followed by the formation of acetate as the key metabolite through the intermediary oxidative and reductive fermentation pathways 9 . Another fate for pyruvate is the conversion of acetyl-CoA, which afterwards enters into the Krebs cycle and generates CO 2 as a by-product 18 . Therefore, the administration of 13 C-glucose (either from 13 C-enriched exogenous glucose or naturally abundant 13 C-glucose) facilitates the production of 13 CO 2 in the by-product CO 2 in presence of H. pylori infection. Thereafter, the major metabolite CO 2 ( 13 CO 2 and 12 CO 2 ) produced by all these metabolic processes is then transported through the blood streams and eventually excreted as 13    In conclusion, our new findings point to a fundamental mechanism underlying both the 18 O and 13 C stable isotopic fractionations of the major metabolite CO 2 in human breath related to glucose metabolism of H. pylori infection in humans. Subsequently, we have taken a step towards unravelling the potential metabolic pathways linking the 18 O and 13 C-isotopic exchange of breath CO 2 and the glucose uptake by H. pylori, thus suggesting that breath 12 C 18 O 16 O and 13 C 16 O 16 O in response to glucose ingestion could be used as potential molecular biomarkers to distinctively track the pathogenesis of H. pylori infection in a non-invasive approach. Although many imperative gaps remain in our understanding of these processes and in the pathophysiology underlying the isotopic exchange and glucose metabolism, our studies may provide new perspectives in the isotope-specific molecular diagnosis of H. pylori infection and hence may pave the way for broad clinical applications along with eradication purposes following standard therapies. Furthermore, new insight into the mechanism linking the isotopic exchange in breath molecule CO 2 to glucose metabolism of H. pylori is fostering exploration of the molecular basis of this infection and new and better approaches together with new pharmacological targets to prevent or treat the deleterious effects of the world's most common gastric pathogen.

Materials and methods
Subjects. Two hundred and twenty four individuals (135 male and 89 female with average age of 39 ± 10 yrs (SD)) were enrolled for this study with different gastrointestinal disorders such as active peptic ulcer disease (PUD), chronic gastritis, and univestigated dyspepsia. We categorized all the human subjects in two distinct groups: infected with H. pylori (H. pylori positive patients: 124) and without the infection of H. pylori (H. pylori negative patients: 100) depending on the reports of gold standard invasive and non-invasive methods, i.e. endoscopy and biopsy based rapid urease test (RUT) and 13 C-urea breath test ( 13 C-UBT).The 13 C-UBT was considered to be indicative of H. pylori positive when δ DOB 13 C (‰) ≥ 3‰ [19][20][21] . There were no mismatches between the two test-reports of all the subjects enrolled in this study (Supplementary Table 1). Exclusion criteria included patients with previous history of diabetes and gastric surgery, taking antibiotics, proton pump inhibitors or H 2 receptor antagonists in the four transporter is phosphorylated to produce glucose-6-phosphate which subsequently incorporated with three potential metabolic routes: glycolysis, pentose phosphate and the Entner-Doudoroff pathway to finally produce major metabolite CO 2 .This major metabolite CO 2 is then transported through the blood streams and eventually excreted as 13  week prior to endoscopy and 13 C-UBT. We received the Ethical approval from the Ethics Committee Review Board of AMRI Hospital, Salt Lake, Kolkata, India (Study no.: AMRI/ETHICS/2013/1). The current protocol has also been approved by the institutional administrative of S. N. Bose Centre, Kolkata, India (Ref. no.: SNB/PER-2-6001/13-14/1769) and the methods were carried out in accordance with the approved guidelines. Informed written consents were taken from all patients participating in this study.
Breath samples collection and measurements. All the human subjects enrolled for the study completed their endoscopic examinations and 13 C-UBTs, 1-2 days prior to glucose breath test (GBT). On the study day before GBT, all the patients were instructed for their mouth-washing to prevent any kind of contact of ingested test meal with the oral cavity bacteria. After an overnight fasting (10-12 hours), an initial baseline breath sample was collected in a 750 ml breath collection bag (QUINTRON, USA, SL No.QT00892) from each subject. After that a test meal of 75 mg U-13 C 6 labelled D-glucose (CIL-CLM-1396-CTM, Cambridge Isotope Laboratories, Inc. USA) or 75 mg unlabeled glucose dissolved in 50 ml water was orally administered to the patient and then subsequent breath samples were collected at 15 minute intervals up to 120 minute. The physical activities of the subjects were restricted inside a room during the test. For the measurements of 18 O/ 16 O and 13 C/ 12 C isotope ratios of exhaled breath CO 2 , a laser-based high-precision ICOS system was employed and the detailed description of the ICOS was given in the following section.
Integrated cavity output spectrometer (ICOS) for breath analysis. For high precision isotopic measurements of breath CO 2 , a high-resolution carbon dioxide analyzer, based on off-axis integrated cavity output spectroscopy (ICOS) method, has been utilized in this study. The detailed description and the measurement accuracy of ICOS method in comparison to the conventional isotope ratio mass spectrometry (IRMS) have been previously demonstrated elsewhere 22,23 . In brief, the laser-based ICOS spectrometer (CCIA 36-EP, Los Gatos research, USA) exploits a high-finesse optical cavity (~59 cm) with two high reflectivity mirrors (R ~ 99.98%) at the both ends of the cavity. This arrangement provides an effective optical path-length of around 3 km through the measuring gas sample, thus offering a high-precision measurement. A continuous wave distributed feedback diode laser operating at ~2.05 μ m is repeatedly tuned over 20 GHz to scan the absorption features of 12 16 O, corresponding to the R (27), P (36) and P (16) ro-virational lines respectively, in the (2,0 0 ,1) ← (0,0 0 ,0) vibrational combinational band of CO 2 , have been utilized to measure the 13 Table 2 and Supplementary Table 3). A 25 mL breath sample was injected into the ICOS cell with a syringe/stopcock for the measurements. High-purity dry nitrogen (HPNG10-1, F-DGSi SAS, France, purity > 99.99%), as the carrier gas, was used to purge the cavity and dilute the breath samples.

Statistical method.
All the data were presented as mean ± SE (Standard Error). For statistical analyses, we performed non-parametric Mann-Whitney test and one way ANNOVA test. A two sided p value < 0.05 was taken account as statistically significant of data. Box-Whiskers plots were utilized to demonstrate the statistical distribution of isotopic enrichments of exhaled breath CO 2 . To obtain the optimal diagnostic cut-off values for δ DOB

18
O‰ and δ DOB 13 C‰ associated with 13 C-labelled ( 13 C-G) and unlabelled glucose ( 12 C-G) metabolism, we performed receiver operating characteristic curve (ROC) analysis (Supplementary Table 4, Supplementary Table 5, Supplementary Table 6 and Supplementary  Table 7). All the data were analysed using Origin Pro 8.0 (Origin Lab Corporation, USA) and Analyse-it Method Evaluation software (Analyse-it Software Ltd, UK, version 2.30).